Tumor Antigens and T-Cell Antigen Recognition Our research focuses on T-cell recognition of antigenic MHC-restricted peptides. The first part of our study addresses how differences on the surface of peptide-MHC complexes can have dramatic effects upon the biology of a responding T-cell clone. Our model system is a xenoreactive murine CD8+ T cell clone (AH3) that recognizes both human HLA-A2 and murine H-2 Db class I MHC when associated with two different peptides. Using a very large panel of peptides, we have shown that AH3 T-cell receptor (TCR) recognition of A2 and Db involves both shared and unique epitopes on the peptide-MHC surface. Furthermore, the AH3 T-cell response to the A2 target does not depend on a CD8-MHC interaction, while its response to the Db target does. The crystal structures of both the A2 and Db MHC ligands of AH3 showed that molecular mimicry does not underlie the cross-reactivity of the AH3 TCR. Subsequently, the structure of human A2-peptide1049 bound to the xenoreactive murine AH3 TCR revealed a non-diagonal binding of the TCR to its cross-reactive MHC ligand. In order to determine the effects of non-diagonal TCR/MHC interactions in a biological setting, we have created an AH3 TCR transgenic mouse to allow us to probe the role of TCR binding orientation in T-cell selection. We have crossed our transgenic mouse with a RAG knockout mouse to generate a strain of mice that express the AH3 TCR exclusively. In addition to the structural differences in the way that the AH3 TCR interacts with its two pMHC targets, there are marked differences in the cytokine profile elicited from AH3 T cells when stimulated with either A2/p1049 or Db/p1058 MHC complexes. To define the underlying causes for this difference, we are using gene microarray and proteomics tools to examine early and late signaling pathways upon stimulation by the two MHC targets. The second part of our study involves T-cell recognition of p53 and Wip1 peptide epitopes in cancer. It is well established that the immune system can participate in an effective anti-tumor response and that tumor rejection in animal models and humans is critically dependent on the generation of antigen-specific T-cell-mediated responses, forming the basis for development of vaccines for cancer immunotherapy. The majority of T-cell-defined human tumor antigens identified to date are widely expressed, shared tumor-associated antigens. For the most part, these epitopes are non-mutated self peptides derived from proteins overexpressed in tumors relative to normal tissues. Three HLA-A2.1-restricted, CTL-defined wild type p53 and Wip1 epitopes have been identified for use in cancer vaccines; they are p5365-73, p53149-157, and p53264-272 and Wip13-11, Wip120-28, Wip1182-190. In recent studies, we found that autologous dendritic cells pulsed with peptide could be used for in vitro restimulation (IVS) of PBMC. Our results indicated, though, that only anti-p53264-272 and anti-Wip13-11 CTLs could be generated from a few normal donors and cancer patients tested. Our goal is to use immunogenic variant peptides of these self wild type p53 and Wip1 epitopes to reverse the non-responsiveness of these patients. In addition, we plan to characterize CD4+ T helper cell-defined wild type p53and Wip1 peptides. This will result in better applicability of a peptide-based vaccine to a larger percentage of the population. In the final part of our study, we addressed the role of Wip1 in T and B cell biology through the use of a Wip1-deficient mouse model. These mice were viable, but showed a variety of postnatal abnormalities. Notably, mice lacking Wip1 showed increased susceptibility to pathogens and diminished T- and B-cell functionality. Previous analysis of both T- and B-cell populations from Wip-1 null animals showed compromised functions. Recently, we found that in Wip1-/- mice, B cell numbers started off low and remained significantly lower throughout the animals life. Both B cell proliferation and apoptosis were similar when compared to wild type mice; however, Wip1-deficient B cells were observed to accumulate at the G2/M phase with reduced cell numbers in S phase after Toll-like receptor (TLR) stimulation. To determine if this defect was occurring at the site of B cell development, we analyzed B cell ontogeny in the bone marrow by phenotype analysis using flow cytometry. Similar numbers of pro and pro/pre B cells were observed irrespective of Wip1 expression; in contrast, IgM-expressing B cells and more mature B cell phenotypes were significantly reduced in number in the bone marrow of Wip1-/- mice, suggesting that Wip1 may also play a role in B cell ontogeny